This invention relates to data processing by digital computer, and more particularly to determining software deployment parameters.
An interconnected collection of computing resources may be referred to as a system landscape. Such a system landscape includes the computers, networks, and other connected devices available within a given scope. The system landscapes of large enterprises are generally complex and heterogeneous. They include various types of computers (such as mainframes, servers or workstations), different storage systems, and multiple network topologies, with various levels of interconnection. Further, the operators of such a system may have a variety of policies and rules governing the use of these resources, designed to optimize such factors as availability, quality of service, and cost.
Modern software applications exhibit a similar degree of complexity, i.e., they have complex interdependencies, different sizing variations, and specific system requirements. This complexity is compounded, generating significant challenges, when such software is to be installed into a target system landscape. Since many dependencies have to be taken into account, and multiple combinations of resources may satisfy the software's requirements to varying degrees of efficiency, it is difficult to manually find an appropriate assembly of all the required resources needed to run the software in a target landscape that ensures optimal operation while complying with the policies of the operators of the system.